A COMPARATIVE ANALYSIS OF HANDICAPPER AND STATISTICAL MODEL SPORTS GAMBLING RETURNS
Open Access
- Author:
- Root, Matthew George
- Area of Honors:
- Finance
- Degree:
- Bachelor of Science
- Document Type:
- Thesis
- Thesis Supervisors:
- Dr. Robert Alexander Novack, Thesis Supervisor
Dr. Brian Spangler Davis, Thesis Honors Advisor - Keywords:
- gambling
sports
finance
statistical modeling
nfl
ncaa
sports gambling - Abstract:
- This thesis tracks game picks from two sports gambling services. The first is a model from Sportsinsights.com, which predicts the winner of an athletic event using statistical and economic concepts. The second is a group of handicappers from Sportsline.com, who use an individual pundit’s intuition and expertise in a particular sport to make their picks. The purpose of this thesis is to determine which type of these services, models or handicappers, offers more profitable picks for potential subscribers. This thesis’ research analyzes picks on games in the NFL, NCAA Men’s Basketball, and NCAA Football. The results are conflicting. Handicappers produced more accurate and profitable picks in NFL games with a 12.68 percent return, compared to a 8.98 percent model return. However, the model outperformed the handicappers in both NCAA Men’s Basketball and Football with returns of -9.85 percent versus -13.35 percent and 2.38 to -12.45 percent respectively. If each service’s picks are considered a portfolio, then the research has determined a model’s picks, made through statistical analysis, is superior to a handicapper’s pick, made by his individual expertise. However, the varying degrees of profitability and inconsistency in success of these professional sites suggest that sports gambling can be a difficult and unreliable profit platform.